You are viewing a plain text version of this content. The canonical link for it is here.

Posted to yarn-issues@hadoop.apache.org by "Jason Lowe (JIRA)" <ji...@apache.org> on 2014/04/08 18:28:23 UTC

[jira] [Commented] (YARN-1913) Cluster logjam when all resources are consumed by AM

    [ https://issues.apache.org/jira/browse/YARN-1913?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13963143#comment-13963143 ] 

Jason Lowe commented on YARN-1913:
----------------------------------

Which scheduler are you using?  The CapacityScheduler already has a yarn.scheduler.capacity.maximum-am-resource-percent property, and there's a per-queue form of it as well.

> Cluster logjam when all resources are consumed by AM
> ----------------------------------------------------
>
>                 Key: YARN-1913
>                 URL: https://issues.apache.org/jira/browse/YARN-1913
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: scheduler
>    Affects Versions: 2.3.0
>            Reporter: bc Wong
>
> It's possible to deadlock a cluster by submitting many applications at once, and have all cluster resources taken up by AMs.
> One solution is for the scheduler to limit resources taken up by AMs, as a percentage of total cluster resources, via a "maxApplicationMasterShare" config.



--
This message was sent by Atlassian JIRA
(v6.2#6252)